HDTwin Product Family

An introduction to the HDTwin product ecosystem — HDTwin Council and HDTwin studio — and a map of the user documentation.

HDTwin is an opinion-gathering platform that replaces traditional market and social research with Human Digital Twins — synthetic personas that replicate a target population and can be interrogated again and again. This page introduces the two products that make up the platform and explains how the rest of the user documentation is organised.

1. HDTwin at a Glance

Traditional research is linear and expensive: define a problem, recruit a panel, run a survey, analyse the results — then start over for the next question. HDTwin turns this into a build once, ask endlessly model.

The idea is simple:

A panel of Human Digital Twins is created once from real research data. Each twin is a synthetic persona that reasons and behaves like the population it represents, not a static record of answers.
That panel can then be used to run any number of deliberations — simulated focus groups and polls in which the twins discuss a topic, vote, and produce an analysed report.

HDTwin delivers faster, scalable, and unbiased insights by moving from one-shot surveys to a reusable population of agents.

2. The HDTwin Product Family

The platform consists of two complementary products, each responsible for one half of the workflow above.

2.1 HDTwin Council

HDTwin Council is the web application in which deliberations happen. A user loads a panel (a .hdt file), defines a topic, and starts a deliberation. The Human Digital Twins in the panel then discuss the topic, vote, and produce a structured report — including consensus and divisiveness analysis — that the user can read and export.

Council is the product you use when you already have a panel and want to run research with it.

2.2 HDTwin studio

HDTwin studio is the framework used to build panels. It is a set of complementary command-line tools, driven by an agentic development environment, that transform source research data (for example survey microdata) into a validated, compiled panel: a collection of synthetic personas packaged as a signed .hdt file.

studio is the product you use when you need to create a panel from research data.

2.3 The fundamental distinction

	HDTwin studio	HDTwin Council
Role	Builds panels	Uses panels
Question it answers	How do I turn research data into a panel of Human Digital Twins?	How do I run a deliberation with a panel and read the results?
Form	CLI tool framework, driven by an agentic environment	Web application
Input	Source research data	A compiled panel (`.hdt`)
Output	A validated panel (`.hdt`)	A deliberation and its report
Typical user	Certified System Integrator	Researcher / end customer

A simple way to remember the split: studio produces the panel, Council consumes it.

3. How Council and studio Relate

The two products are not separate offerings so much as two stages of a single workflow, connected by one shared artifact: the panel.

@startuml
skinparam componentStyle rectangle

rectangle "Source research data" as SRD
rectangle "Panel (.hdt)" as PN
rectangle "Deliberation" as DEL
rectangle "Report" as REP

SRD --> PN : HDTwin studio (build)
PN --> DEL : HDTwin Council (use)
DEL --> REP
@enduml

studio sits at the start of the pipeline. It takes real research data and reconstructs it into a population of synthetic personas. The result is compiled into a portable, signed .hdt file.
Council sits at the end. It takes that same .hdt file and uses it to run deliberations.

The .hdt file is the contract between the two products: whatever studio produces, Council can run. A panel is built once with studio, then used many times in Council — each new research question becomes a new deliberation on the same panel, with no re-recruitment.

The two products also share the same underlying platform. Both rely on HDTwin-optimised language models reached through a common gateway, and Council’s deliberations are executed by a dedicated deliberation engine (the Habermas Machine) configured inside each panel.

Conceptually, then, HDTwin is one platform split across two responsibilities — panel construction and panel deliberation — with the panel as the hand-off point between them.

4. Choosing the Right Guide

Which part of the documentation you need depends on what you want to accomplish:

I want to…	Read…
Run a deliberation: load a panel, start a discussion, read and export the report	The HDTwin Council application guide — Deliberation
Build a panel: go from research data to a validated, delivered `.hdt`	The HDTwin studio guide — System Integrator Guide
Understand the concepts behind panels, personas, and reconstruction	The studio guide first; it explains the conceptual model as part of the workflow
Look up a term (panel, persona, LCM, `.hdt`…)	The Glossary
Check legal and compliance obligations	The Compliance section

If you are an end researcher who has received a .hdt file and simply wants to gather opinions, start with the Deliberation guide. If your job is to create that file in the first place, start with the System Integrator Guide.

5. Documentation Map

The user documentation is small and organised by product. Each guide is self-contained, but they are best read in the following order:

This page — the front door: what HDTwin is, and how Council and studio fit together.
Deliberation — the HDTwin Council application guide. Covers loading a panel, starting a deliberation, and reading the report.
HDTwin studio — the System Integrator guide. Covers the end-to-end workflow for building a panel, from source data to a validated .hdt.
Glossary — the terms used across the documentation, defined in one place.

Beyond the user guides, the documentation also contains a Compliance section (GDPR, EU AI Act, ethics) that applies to all uses of the platform.